PLS-GEP algorithm-based air quality prediction

Air pollution is an increasing threat to human health and the environment, and accurate air quality prediction has become a research priority. In this study, a novel algorithm based on Partial Least Squares - Gene Expression Programming (PLS-GEP) is proposed and its performance is compared with that of support vector machine (SVM) and long-short-term memory network (LSTM). The model is evaluated in terms of MSE, RMSE, MAPE and ACC using day-by-day air quality data in Beijing, respectively. The experimental results show that PLS-GEP significantly outperforms other algorithms in both prediction accuracy and classification ability, with a classification accuracy of 92.75%. The study shows that the PLS-GEP algorithm has obvious advantages in air quality prediction tasks and provides a reliable method for improving prediction accuracy and decision support.